CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Methods: HIV+/HCV+ participating in the Biomarker Study who reported having had a drink in the past 12 months or had a detectable PEth were included (n=663). We restricted analysis to current drinkers to avoid confounding from those who stopped due to illness. Because any alcohol use may have health effects among HIV+/HCV+, any detectable value of PEth (>0 ng/mL) was considered indicative of exposure and a threshold of 8+ ng/mL as harmful. PEth thresholds were compared with self report of “any use over 12 months” and an AUDIT-C score of 4+ for men (3+ for women) indicated harmful use. We used separate Cox regression to estimate mortality hazard ratios by exposure using PEth (0, 1-7, 8-49, 50-99, 100+ ng/mL) and AUDIT-C (0, <4, 4-5, 6-7, 8+). Models were adjusted for age, race, and disagreement between measures. Results: Agreement for any exposure was 59%; despite a shorter time frame (14 days), detectable PEth classified 91% of the sample as exposed compared with 68% by self report of use in the past year. When PEth 8+ ng/mL (45%) was compared with AUDIT-C for harmful use (29%), agreement improved to 72%. Both PEth and AUDIT-C were associated with mortality (Figure below), but PEth demonstrated stronger associations throughout its range. Finally, self-report did not enhance the association between alcohol and mortality beyond that identified by PEth (p>0.2). Conclusions: Among HIV+/HCV+, PEth is more sensitive than AUDIT–C for detecting alcohol exposure associated with mortality and thresholds for clinically important exposure may need to be reduced. Based on a dried blood spot, PEth offers a practical, direct measure of exposure.

555 HLA-B18 As a Risk Factor of Progression to Severe Liver Fibrosis in HIV/HCV Patients Mario Frías 1 ; Antonio Rivero-Juárez 2 ; Francisca Cuenca 3 ; Diego Rodríguez-Cano 2 ; Juan Macías 4 ; Ana Gordon 2 ; Ángela Camacho 2 ; Juan A. Pineda 4 ; Antonio Rivero 3 1 Instituo Maimonides de Investigación Biomédica de Córdoba, Córdoba, Spain; 2 IMIBIC, Córdoba, Spain; 3 Hosp Universitario Reina Sofía, Córdoba, Spain; 4 Hosp Universitario de Valme, Sevilla, Spain Background: Chronic infection by hepatitis C virus (HCV) is characterized by progressive liver fibrosis. Although several factors have been found to be associated with faster liver fibrosis progression, a biomarker with sufficient predictive value to impact clinical decision making has not been identified. The aim of our study was to analyze the influence of HLA-B molecules on liver fibrosis progression in HIV/HCV patients without cirrhosis. Methods: A retrospective longitudinal study included HIV/HCV patients in follow-up between 2007 and 2014. Inclusion criteria were: a) at least two determinations of Liver Stiffness Measurement (LSM); b) ≥12 months of total follow-up; c) active HCV infection. Exclusion criteria were: a) HCV treatment previous to follow-up or during the first 12 months of the study; b) cirrhosis at baseline, defined as an LSM≥14.6 kPa. Analytical, demographic and clinical variables were collected. LSM cutoffs applied were: <6.5 kPa (F0-F1); 6.5-9.4 kPa (F2); 9.5-14.5 kPa (F3) and ≥14.6 kPa (F4). Outcome variables were: 1) fibrosis stage progressed at least one stage; 2) fibrosis progressed up to severe liver fibrosis (F3-F4). Patients were censored at: a) an event of interest; b) initiation of HCV treatment; c) loss of follow-up; d) end date of study. Results: One-hundred and four patients constituted the study population. The patient distribution according to liver fibrosis was as follows: F0-F1, 62 (59.6%) patients; F2, 22 (21.2%) patients; and F3, 20 (19.2%) patients. The median (IQR) follow-up period for these patients was 54.5 (IQR: 26.2-77) months. Forty-five patients (43.3%) showed an increase in the degree of fibrosis (time to event: 29 [IQR: 14-49.5] months). Patients bearing HLA-B18 more frequently had liver fibrosis and had a faster liver fibrosis progression rate (73.3%; time to event 24 [IQR: 8-29] months) than HLA-B18neg patients (38.2%; time to event 34.5 [IQR: 14.7-51.2] months). This association was also observed in the development of F3-F4 among F0-F2 patients (HLA-B18pos: 69.2%; time to event = 18 [6.5-37] months vs. HLA-B18neg: 28.2%; time to event = 37 [IQR: 19-52] months). Conclusions: HIV/HCV patients carrying allele HLA-B18 were more likely to progress more rapidly to developing advanced and severe fibrosis (F3-F4) than HLA-B18neg patients. These results could help make decisions about the timing of HCV therapy in F0-F2 patients at risk of accelerated progression. 556 Impact of PNPLA3 Variants on the Liver Histology of 168 HIV/HCV Patients Caterina Sagnelli 1 ; Marco Merli 2 ; Caterina Uberti Foppa 3 ; Hamid Hasson 4 ; Grazia Cirillo 1 ; Anna Grandone 1 ; Stefania Salpietro 4 ; Emanuela Messina 4 ; Carmine Minichini 1 ; Emanuele Miraglia Del Giudice 1 ; Adriano Lazzarin 4 ; Nicola Coppola 1 ; Evangelista Sagnelli 1 1 Second Univ of Naples, Naples, Italy; 2 Vita-Salute San Raffaele Univ, Milan, Italy; 3 San Raffaele Scientific Inst, Milan, Italy; 4 Vita-Salute Univ, San Raffaele Scientific Inst, Milan, Italy Background: A genome-wide association study recently identified a non-synonymous sequence variation (rs738409 C to G) encoding an isoleucine-to-methionine substitution at position 148 in the adiponutrin/patatin-like phospholipase-3 (PNPLA3) gene, which appears to be the strongest determinant of human steatosis.The I148M variant of PNPLA3 has been found to be independently associated with liver steatosis and disease progression in NAFLD , alcoholic liver disease, CHC and HCC. Methods: This study analyzes the impact of PNPLA3 variants on liver histology of 168 HIV/HCV coinfected patients naïve for HCV-treatment. A pathologist unaware of the patients’ condition graded liver fibrosis and necroinflammation (Ishak) and steatosis (Kleiner). Patients were tested for PNPLA3 variants and genotyped for the PNPLA3 rs738409 C to G variant underlying the I148M substitution. All were HBsAg-negative and stated no alcohol abuse. Results: The mean age was 40.6(37.6-44.1), 72.6%were males, 42% showed HCV-genotype 3, 38.9% HCV-genotype 1 and 79.2%were receiving HAART. The 79 patients with the PNPLA3 p.148I/M or M/M variants more frequently showed severe steatosis (score 3-4) than the 89 with PNPLA3 p.148I/I, 43% vs. 24.7%( p =0.001), whereas no difference was observed in the degree of necroinflammation or fibrosis. Compared with 112 patients with lower scores, 56 with severe steatosis showed higher BMI ( p =0.03), higher rate of HCV- genotype 3 (55.6% vs. 35.2%, p =0.01) and PNPLA3p.148I/M or M/M (60.7% vs. 39.3%, p =0.01) and lower CD4+ cells/mm3 [514.00(390.5-673.0) vs. 500.00(399.0-627.0]; p =0.002). At multivariate analysis, BMI ( p =0.01), HCV-genotype 3 ( p =0.006), CD4+ cell count (p =0.005) and the PNPLA3 p.148I/M or M/M variants ( p =0.01) were found to be independent predictors of severe liver steatosis. Besides, the PNPLA3p.148 I/M or M/M variants and CD4+ cell count were the only independent predictors of severe steatosis in patients with HCV-genotype non-3. Conclusions: This is the first study to show that among patients HIV/HCV coinfected patients the PNPLA3 p.148I/M or M/M variants has substantially less impact on steatosis in those with HCV-genotype 3 than in non-genotype 3 infected patients.

Poster Abstracts

219

CROI 2016